GHSA-MJ2C-8HXF-FFVQ
Vulnerability from github – Published: 2025-05-19 16:25 – Updated: 2025-05-19 16:25Summary
A command echoing feature in the framework allows users to indirectly trigger privileged behavior by injecting special platform tags. Specifically, an unauthorized user can use the /echo <qqbot-at-everyone /> command to cause the bot to send a message that mentions all members in the chat, bypassing any permission controls. This can lead to spam, disruption, or abuse of notification systems.
Details
The framework provides a command /echo that causes the bot to repeat any user-provided message verbatim in the group chat. However, the bot fails to sanitize or filter platform-specific control elements such as <qqbot-at-everyone />, which, when included in a message, mentions everyone (i.e., @全体成员). While normal users are forbidden from using this tag in normal chats, the bot, which has higher privileges, is allowed to do so.
Since the /echo command blindly echoes any content, a user can exploit this by sending:
/echo <qqbot-at-everyone />
The bot will then send a message containing <qqbot-at-everyone />, causing the platform to interpret it as an @全体成员 command, effectively allowing an unauthorized user to @everyone via the bot.
PoC
1. Set up a chatbot using the affected framework. 2. Join the chat that includes the bot as a regular user with no permission to use @全体成员. 3. Send the following message in the chat:
/echo <qqbot-at-everyone />
- The bot will respond by repeating the message, and the platform will interpret
<qqbot-at-everyone />as an @全体成员 mention. - All the chat members receive a notification, despite the user lacking that permission.
{
"affected": [
{
"package": {
"ecosystem": "npm",
"name": "cocotais-bot"
},
"ranges": [
{
"events": [
{
"introduced": "1.5.0-test2-hotfix"
},
{
"fixed": "1.6.2"
}
],
"type": "ECOSYSTEM"
}
]
}
],
"aliases": [
"CVE-2025-47948"
],
"database_specific": {
"cwe_ids": [
"CWE-74"
],
"github_reviewed": true,
"github_reviewed_at": "2025-05-19T16:25:05Z",
"nvd_published_at": "2025-05-17T19:15:46Z",
"severity": "MODERATE"
},
"details": "### Summary\n\ufeff\nA command echoing feature in the framework allows users to indirectly trigger privileged behavior by injecting special platform tags. Specifically, an unauthorized user can use the `/echo \u003cqqbot-at-everyone /\u003e` command to cause the bot to send a message that mentions all members in the chat, bypassing any permission controls. This can lead to spam, disruption, or abuse of notification systems.\n\ufeff\n### Details\n\ufeff\nThe framework provides a command `/echo` that causes the bot to repeat any user-provided message verbatim in the group chat. However, the bot fails to sanitize or filter platform-specific control elements such as `\u003cqqbot-at-everyone /\u003e`, which, when included in a message, mentions everyone (i.e., @\u5168\u4f53\u6210\u5458). While normal users are forbidden from using this tag in normal chats, the bot, which has higher privileges, is allowed to do so.\n\ufeff\nSince the `/echo` command blindly echoes any content, a user can exploit this by sending:\n\ufeff\n```\n/echo \u003cqqbot-at-everyone /\u003e\n```\n\ufeff\nThe bot will then send a message containing `\u003cqqbot-at-everyone /\u003e`, causing the platform to interpret it as an @\u5168\u4f53\u6210\u5458 command, effectively allowing an unauthorized user to @everyone via the bot.\n\ufeff\n### PoC\n\ufeff\n1. Set up a chatbot using the affected framework.\n2. Join the chat that includes the bot as a regular user with **no permission to use @\u5168\u4f53\u6210\u5458**.\n3. Send the following message in the chat:\n\ufeff\n```\n/echo \u003cqqbot-at-everyone /\u003e\n```\n4. The bot will respond by repeating the message, and the platform will interpret `\u003cqqbot-at-everyone /\u003e` as an @\u5168\u4f53\u6210\u5458 mention.\n5. All the chat members receive a notification, despite the user lacking that permission.\n\ufeff\n\ufeff",
"id": "GHSA-mj2c-8hxf-ffvq",
"modified": "2025-05-19T16:25:05Z",
"published": "2025-05-19T16:25:05Z",
"references": [
{
"type": "WEB",
"url": "https://github.com/cocotais/cocotais-bot/security/advisories/GHSA-mj2c-8hxf-ffvq"
},
{
"type": "ADVISORY",
"url": "https://nvd.nist.gov/vuln/detail/CVE-2025-47948"
},
{
"type": "WEB",
"url": "https://github.com/cocotais/cocotais-bot/commit/d1cf01a9a41b3131241d1833444b890c8d6e70b8"
},
{
"type": "PACKAGE",
"url": "https://github.com/cocotais/cocotais-bot"
}
],
"schema_version": "1.4.0",
"severity": [
{
"score": "CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:L/A:L",
"type": "CVSS_V3"
}
],
"summary": "Cocotais Bot has builtin .echo command injection"
}
Sightings
| Author | Source | Type | Date | Other |
|---|
Nomenclature
- Seen: The vulnerability was mentioned, discussed, or observed by the user.
- Confirmed: The vulnerability has been validated from an analyst's perspective.
- Published Proof of Concept: A public proof of concept is available for this vulnerability.
- Exploited: The vulnerability was observed as exploited by the user who reported the sighting.
- Patched: The vulnerability was observed as successfully patched by the user who reported the sighting.
- Not exploited: The vulnerability was not observed as exploited by the user who reported the sighting.
- Not confirmed: The user expressed doubt about the validity of the vulnerability.
- Not patched: The vulnerability was not observed as successfully patched by the user who reported the sighting.
The approach is described in our paper Mapping CVEs to MITRE ATT&CK Techniques: A Curated Gold-Set Classifier and the Limits of LLM-Assisted Label Expansion.
Browse all ATT&CK techniques and the vulnerabilities related to each.
Related by attack behaviour
Vulnerabilities whose description is nearest to this one in the vector space of the CIRCL/vulnerability-attack-technique-biencoder model. This is a similarity search over the bi-encoder space (plain cosine), not a classification, and it has no measured accuracy.